Summary paragraph: Our section directs clinical trials of recombinant toxins at the NIH and in the lab studies the interaction of such toxins with the patient's malignant and normal tissues. Therapy of patients with chemotherapy resistant hematologic malignancies has been remarkably effective with Fv-toxins (recombinant immunotoxins) targeting CD25 or CD22 in hairy cell leukemia and efficacy is observed in some other hematologic malignancies as well. The anti-CD22 recombinant immunotoxin BL22 produces complete remissions (CRs) in the majority of patients with hairy cell leukemia who have little chance for CR with conventional therapies. We are continuing to treat patients to optimize safety and efficacy. Patient samples are studied in the lab to better predict and understand recombinant toxin efficacy, to better understand the role of cytotoxic T-cells observed in most patients receiving BL22, and to study novel and established tumor markers in patients to better quantitate overall tumor burden. Our goal is to help develop bacterial toxins which are engineered to kill cancer cells in patients who cannot be cured by standard therapy. The binding domain of the bacterial toxin, usually Pseudomonas exotoxin (PE), is replaced with a ligand, either a growth factor or an Fv fragment of an monoclonal antibody (MAb), which binds to a tumor associated antigen. Thus the recombinant toxin binds to and is internalized by malignant cells and causes cell death after the catalytic domain of the toxin enters the cytosol of the host cell. BL22 is a recombinant immunotoxin containing an anti-CD22 Fv fused to truncated PE. We had previously shown in the lab that BL22 was capable of inducing complete regressions of human CD22+ tumor xenografts in mice at plasma concentrations tolerated in monkeys and in killing malignant cells freshly obtained from patients with CD22+ B-cell leukemias. We completed accrual of 46 patients in a phase I trial of BL22 in patients with B-cell tumors, 31 of whom had hairy cell leukemia (HCL). Of these 31 HCL patients, 19 (61%) achieved CR and 6 (19%) had partial responses (PR). Cytopenias resolved in all responders. Four HCL patients on phase I and an additional HCL patient receiving BL22 by special exemption had a reversible hemolytic uremic syndrome (HUS). All 5 HUS patients with HCL recovered from HUS with normal renal function and responded to BL22 with marked improvements in depressed normal blood counts. Thus BL22 has an excellent risk-benefit ratio even in patients who had HUS. A phase II trial is currently underway to establish its safety and efficacy in HCL. In studying the interaction of CD22 from patients' cells with BL22, it was noted for the first time that soluble CD22 could be detected in the serum and its level correlates closely with overall disease burden. Another phenomenon studied is expansions of cytotoxic T-lymphocytes (CTLs) in most HCL patients treated with BL22. These CTLs are currently being characterized in the lab to determine if they are reacting with HCL antigens to benefit clinical responses, and to determine whether increased T-cell immunity after BL22 therapy might be advantageous compared with purine analog chemotherapy which damages T-cells. To target the mesothelin antigen on mesotheliomas, ovarian carcinoma, and other squamous cancers, the recombinant immunotoxin SS1P was developed in Dr. Pastan's lab and is currently being tested in patients with mesothelin- expressing malignancies. To achieve better tumor penetration the recombinant toxin is being administered by continuous infusion. We have currently reached the 5th dose level with encouraging results and are continuing to enroll patients to determine the maximum tolerated dose. Solid tumors present important challenges to the efficacy of recombinant toxins, including tumor penetration and immunogenicity. Approaches which are being developed to try to overcome these challenges may be useful in the treatment of patients with both solid and hematologic tumors lacking effective alternative therapies.